Summary of Project 1: The Role of Physical Cues in Collective Cell Invasion The ability of tumors to invade adjacent tissues, leading to local or distant metastasis, is a hallmark of cancer. Cancer cells frequently invade as groups of adherent cells in a process termed collective invasion. Previous studies have primarily focused on single cell or semi-collective (multicellular streaming) cell invasion. Single cell models for metastasis have direct implications for tumors whose cells migrate constitutively as individual cells, such as leukemias and lymphomas, or after cell detachment from a primary tumor via epithelial-to- mesenchymal transition (EMT). However, EMT has long been controversial among pathologists as breast tumors at metastatic sites typically display epithelial features. While EMT-like gene signatures can be observed in specific mouse models and breast cancer subtypes, the majority of breast tumors do not exhibit clear molecular features of EMT. Intravital microscopy studies reveal that tumor cells preferentially migrate collectively along pre-existing channels that are defined by various anatomical structures in vivo. However, it is currently unknown how the physical properties of the microenvironment, such as confinement and compliance, regulate the molecular mechanisms of collective cell invasion. Intriguing preliminary data reveal that cancer cells migrate through wide (?50 m) tracks as a collective unit. However, as confinement increases, the cancer cells spontaneously disseminate, first as clusters of 2-5 cells and eventually, in very narrow tracks (?10 m), as single cells. We hypothesize that the physical microenvironment induces a signaling cascade of events that transforms the classical collective to single cell invasion. To test this hypothesis, we will employ a multidisciplinary approach combining novel bioengineering tools and mathematical modeling with sophisticated molecular cell biology and imaging techniques and in vivo models. In Aim 1, we will develop an integrated experimental and computational model of collective cell movement in confined geometries modeling primary tumor invasion, and dissect the mechanisms by which cell-cell contact is released during mechanically-induced transitions to single cell movement, focusing on the role of E-cadherin cleavage and possible EMT induction. In Aim 2, we will delineate the relative contributions of actomyosin contractility, small GTPases and osmotic engine model to locomotion in rigid versus compliant confined microenvironments. In Aim 3, we will validate our in vitro understanding of the dissemination and locomotion of cancer cells in more complex microenvironments characteristic of in vivo breast tumors using an organotypic 3D culture system and genetically engineered mouse models. Elucidation of the underlying mechanisms of collective cancer cell invasion will offer insights into our understanding of how cancer cells spread through the body, and it could shift the currently prevailing single cell paradigm in cancer to incorporate concepts of mechanical signaling, cell-cell adhesion, and cell-cell cooperation.